Abstract
BACKGROUND: Sub-Saharan African countries remain adversely affected by the burden of liver cirrhosis, recording the highest age-standardized death rates in the world. Hospital admission represents a measurable proxy of the burden of cirrhosis, driving up healthcare costs and imposing a burden on patients and caretakers. This study was set out to fill a local gap in knowledge of the rate and predictors of readmission among patients with decompensated cirrhosis which is crucial in order to reduce it. METHODS: We conducted a hospital-based prospective cohort study in the Gastroenterology unit of the Muhimbili National Hospital in Tanzania. Applying consecutive sampling, we recruited 110 consenting patients with decompensated liver cirrhosis 18 years and older admitted between August and November 2023 and followed them up for 90 days after hospital discharge. Patients who underwent transplant, died during the index admission and those discharged against medical advice were excluded. Socio-demographic, clinical and laboratory data were recorded using a structured questionnaire. Follow up post-discharge was done through phone calls. Data analysis was done using STATA version 15.1 (StataCorp, Texas, USA). Continuous variables were summarized as medians with interquartile ranges and categorical variables as counts with percentages. Comparison was done using the Kruskal–Wallis test for continuous variables and Chi-squared test and Fisher exact test for categorical variables. The Robust Poisson regression model was used to analyze for predictors of readmission at 1 and 3 months. A p-value of 0.05 was used to denote statistical significance. RESULTS: Of 110 patients, 71.8% were male. Median age was 44.5. Leading etiology of cirrhosis was alcohol (38.2%), leading decompensating event responsible for index hospitalization was ascites (68.2%). One-month and three-month readmission rates were 20.9% and 30.9% respectively. Predictors of readmission at 1 month were Model for End-stage Liver Disease- Sodium (MELD-Na) score (p = 0.029) and at 3 months were MELD-Na score (p = 0.006) and thrombocytopenia (p = 0.023). CONCLUSION: We found a high readmission rate among patients with decompensated cirrhosis, similar to literature from other countries. The identification of the MELD-Na score and platelet count as independent predictors of readmission provides valuable insight into risk factors for poor outcomes among patients with cirrhosis.